Correlation vs. Causation

Analyst examining two parallel health trends separated from a possible causal pathway.

Correlation vs. Causation Explainer

Correlation means that two variables vary together. Causation means that changing one would produce a change in the other under the relevant conditions. A correlation can be useful evidence, but by itself it does not establish which variable influences the other or whether either is responsible for the pattern.

A third factor may affect both variables. For example, age can be associated with an exposure and a health outcome, creating a relationship that changes after age is considered. Reverse causation is another possibility: early disease may alter behavior, making the behavior appear to precede illness when the direction is partly reversed.

Chance, selection, measurement error, and shared time trends can also produce misleading correlations. Two population averages may move together even when the corresponding individual-level relationship is different, a problem known as ecological fallacy. A strong correlation is therefore not a substitute for a suitable design.

Causal assessment looks at temporality, comparison groups, confounding control, dose–response patterns, experimental or quasi-experimental evidence, mechanisms, and consistency across studies. No single checklist item proves causation. The conclusion comes from how well the total evidence rules out credible alternative explanations.

A third variable may influence both measures, the outcome may influence the exposure, or chance may create the pattern.

Each possibility requires a different design or analysis to investigate.

A cause must precede its effect, although long latency can make the relevant time window difficult to define.

Measurements taken at one time point often cannot establish direction.

Randomization, natural experiments, repeated longitudinal evidence, and well-controlled comparisons can reduce alternative explanations.

Biological plausibility and consistency help, but they do not rescue a badly biased design.

Explore more "Explainers"

Discover additional explainers across politics, science, business, technology, and other fields. Each explainer breaks down a complex idea into clear, everyday language—helping you better understand how major concepts, systems, and debates shape the world around us.